Fig. 1

Conceptual framework of HPV testing accuracy for CIN2 +. Panel A reports the base case. The up-left table represents the analytical accuracy, i.e. the ability of the test to correctly identify the virus, In the own-central table we represent the association between HPV infection and CIN2 + in women without HIV using data approximated from the Dreyer et al. 2023 study (7). We considered HPV a necessary cause. The up-right table represents the clinical accuracy of the HPV test for CIN2 + as predicted by the conceptual model. Panel B reports a theoretical population with an increased prevalence of CIN2 + due only to an increase in the prevalence of HPV infection, i.e. the biological link does not change. In this case, clinical positive predictive value (PPV) only slightly increases, while clinical specificity decreases. Panel C reports a population in which the increase in prevalence is due both to an increase in infections and to an increase in the probability of progression from infection to CIN2 +, i.e. the biological link changes. This scenario may be representative of the women living with HIV. In this scenario, clinical PPV increases and clinical specificity decreases